Unpaired electron spin density in biological macromolecules can significantly limit the application of NMR as an independent structural tool, even when other parameters (solubility, size are favorable for NMR structural studies. Although methods have been developed which in many cases allow identification of resonances which are shifted and/or broadened by hyperfine interactions, it is often difficult to obtain complete sequential resonance assignments for proteins which incorporate paramagnetic centers. Recently, we have reported extensive sequential 1H resonance assignments and described the secondary structural elements of putidaredoxin (Pdx), a 2-Fe S-2-containing ferredoxin isolated from Pseudomonas putida. Although progress has been made in refining the solution structure of oxidized Pdx (Fe+3-Fe+3), broadening of 1H resonances due to proximity to the metal cluster results in a considerable loss of spectral information. with ca. 14% of all proton resonances i n oxidized Pdx unobservable due to hyperfine interactions. The metal cluster is required for folding, as the apoprotein is virtually structureless. Recently, the zinc reconstitution of rubredoxin, a protein which normally contains a single iron ion tetrahedrally coordinated by four cysteinyl sulfur ligands, has been reported. However, the presence of two "inorganic" sulfide ions which bridge the iron atoms in a 2-Fe S-2 cluster introduces a further complication. To our knowledge, no Fe2S2 ferredoxins have been successfully reconstituted with non-native metal ions. We are presently shown that Ga+3 is useful for reconstitution of Pdx, giving a folded protein which is structurally similar to the native (ion-containing) form. Mass spectrometry is providing an important probe of the composition of the protein gallium complexes. A paper describing this work appeared in J. Am. Chem. Soc. 117 (1995) 6625. Experiments are planned to investigate protein Zn complexes. During the course of our experiments with gallium putidaredoxin, we di scovered that the molecule is co-valently modified at the thiol functionalities. The sites of modification were established and the source of the modification found to be the mercaptoethanol that was used to ensure a reducing environment. Changing the reducing agent to dithiothreitol got rid of the unwanted modification and opened up the way to successful high resolution X-ray crystallography of certain ferredoxins whose structure had previously been refractory to X-ray analysis.